Date of Award

Winter 2007

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Materials and Infrastructure Systems

First Advisor

David Hall

Abstract

Sewer pipelines are prevalent, important, valuable, unnoticed, and often in a state of disrepair. Pipeline inspection is essential for effective management of wastewater systems and is now mandated for many municipalities complying with the Governmental Accounting Standards Board Statement 34 and EPA regulations. Pipe inspection robots are routinely used to inspect underground pipelines for cracks, deformations, leaks, blockages and other anomalies to prevent catastrophic failure and to ensure cost effective maintenance and renewal. Most existing pipe inspection robots only collect video footage of pipe condition. Pipe profiling technology has recently been introduced to allow for measurement of the internal coordinate geometry of pipelines. Accurate radial measurements permit the calculation of several important pipe parameters which aid in the determination of pipe condition and prediction of time to failure. Significant research work has been completed in North America, Europe, Asia and Australia aimed at improving the accuracy and automation of the pipe inspection process. However, standard calibration, verification, reporting and analysis practices must be developed for pipe profilers if coordinate profiling data is to be effectively included in the long term management of pipeline assets.

The objective of this research is to quantify the measurement error incurred by a pipe profiler's misalignment with the pipe axis, present a new methodology to correct the measurement error, develop a prototype profiler to verify the equations derived herein, and to further the development of pipe profiler technology at the Trenchless Technology Center at Louisiana Tech University. Equations are derived for pipe ovality as a function of the robot's position and orientation with respect to a pipe to demonstrate the magnitude of the error which is introduced by a robot's misalignment with the pipe axis. A new technique is presented to estimate the position and orientation of a profiler using radial measurement devices at each of its ends. This technique is demonstrated by applying homogeneous coordinate transformations to simulated radial measurements based on mathematically generated data that would be obtained by incrementally rotating two parallel radial measuring devices in a perfectly cylindrical pipe. A prototype pipe profiling robot was developed to demonstrate the new position and orientation technique and to experimentally verify the measurement error caused by a robot's misalignment with the pipe axis. This work improves the accuracy and automation of pipe profiling technology and makes a case for the development of industry standard calibration, verification, reporting and analysis practices.

Download

Share

COinS